Associations between end-of-life discussions, patient mental health, medical care near death, and caregiver bereavement adjustment

Abstract

Context

Talking about death can be difficult. Without evidence that end-of-life (EOL) discussions improve patient outcomes, physicians must balance their desire to honor patient autonomy against a concern that EOL discussions may inflict psychological harm.

Objective

To determine whether EOL discussions with physicians are associated with fewer aggressive interventions.

Design, Setting, Participants

A United States multi-site, prospective, longitudinal cohort study of advanced cancer patients and their informal caregivers (n=332 dyads), September 2002-February 2008. Patients were followed from enrollment to death a median of 4.4 months later. Bereaved caregivers’ psychiatric illness and quality of life (QoL) was assessed a median 6.5 months later.

Main Outcome Measures

The primary outcome were aggressive medical care (e.g., ventilation, resuscitation) and hospice in the final week of life. Secondary outcomes included patients’ mental health and caregivers’ bereavement adjustment.

Results

123 of 332 (37.0%) patients reported EOL discussions before baseline. EOL discussions were not associated with higher rates of Major Depressive Disorder (8.3% vs. 5.8; AOR 1.33, 95% CI 0.54-3.32), or more “worry” (6.5 vs. 7.0; p=0.19)). After propensity-score weighted adjustment, EOL discussions were associated with lower rates of ventilation (1.6% vs. 11.0%; AOR 0.26, 95% CI 0.08-0.83), resuscitation (0.8% vs. 6.7%; AOR 0.16, 95% CI 0.03-0.80), ICU admission (4.1 vs. 12.4%; AOR 0.35, 95% CI 0.14-0.90), and earlier hospice enrollment (65.6% vs. 44.5%; AOR 1.58, 95% CI 1.04-2.63). In adjusted analyses, more aggressive medical care was associated with worse patient QoL (6.4 vs. 4.6; F=3.60, p=0.01) and higher risk for Major Depressive Disorder in bereaved caregivers (AOR 3.37, 95% CI 1.12-10.13), while longer hospice stays were associated with better patient QoL (5.6 vs. 6.9; F=3.70, p=0.01). Better patient QoL was associated with better caregiver QoL at follow-up (β=0.20; p=0.001).

Conclusion

EOL discussions are associated with less aggressive medical care near death and earlier hospice referrals. Aggressive care is associated with worse patient QoL and worse bereavement adjustment.

INTRODUCTION

End-of-life (EOL) discussions offer patients the opportunity to define their goals and expectations for the medical care that they want to receive near death. But, these discussions also mean confronting the limitations of medical treatments and the reality that life is finite, both of which may cause psychological distress.¹ Studies suggest that doctors and patients are ambivalent about talking about death, and often avoid EOL conversations.^2-13 To date, however, research has not examined whether these discussions are associated with patients’ psychological distress or medical care near death. Without this information doctors cannot weigh the risks and benefits of EOL discussions.

A decade ago Weeks et al.¹⁴ demonstrated that advanced cancer patients who preferred life-extending therapy were often overly optimistic about their chances of survival. These results suggested that EOL discussions might reduce the aggressiveness of medical care near death by making patients more realistic about the benefits of intensive therapies. Consistent with this, a small study showed that EOL conversations significantly influenced patients’ understanding that their illness was terminal and that this, in turn, was associated with increased integration of hospice.¹⁵ Another study demonstrated that patients who wanted to avoid cardiopulmonary resuscitation and discussed their preferences with physicians were more likely to have do-not-resuscitate orders.¹⁶ Others have shown that hospice is associated with better quality of life (QoL) near death,¹⁷ and reductions in bereaved caregivers’ risk of developing Major Depressive Disorder.¹⁸ To date, however, research has not examined whether EOL discussions are associated with patients’ medical care in the final week of life, or whether the medical care received at the EOL is correlated with differences in patients’ QoL near death or caregivers’ bereavement outcomes.

The primary aim of this prospective, longitudinal cohort study was to examine the associations between EOL discussions with physicians and the medical care that terminally ill cancer patients receive near death. Our a priori hypothesis was that EOL discussions would be associated with less aggressive medical care and earlier integration of hospice. Our secondary aims were to examine the associations between EOL discussions and patients’ psychological distress, QoL near death, and caregivers’ bereavement outcomes. We hypothesized that EOL discussions would be associated with greater psychological distress. Although we did not expect patients’ QoL near death to be directly influenced by EOL discussions reported at baseline, we did expect that patients’ QoL would be influenced by the medical care that they actually received near death. We also hypothesized that patients’ EOL experiences (i.e., their medical care and QoL) would affect the surviving caregivers’ bereavement adjustment.

METHODS

Study sample

Coping with Cancer (CwC) was an NCI- and NIMH-funded prospective, longitudinal, multi-site cohort study of terminally ill cancer patients and their informal caregivers (e.g., spouse or adult child) followed through bereavement. CwC sought to examine how psychosocial factors influencing patients’ care and their caregivers’ bereavement adjustment.

Patients were recruited from September 1, 2002 to February 28, 2008 from seven different outpatient sites: Yale Cancer Center (New Haven, CT), the Veterans Affairs Connecticut Healthcare System Comprehensive Cancer Clinics (West Haven, CT), the Parkland Hospital Palliative Care Service (Dallas, TX), Simmons Comprehensive Cancer Center (Dallas, TX), Massachusetts General Hospital (Boston, MA), Dana-Farber Cancer Institute (Boston, MA), and New Hampshire Oncology-Hematology (Hookset, NH). Approval was obtained from the human subjects committees of all participating centers; all enrolled patients provided written informed consent.

Eligibility criteria included: diagnosis of advanced cancer (presence of distant metastases and disease refractory to first-line chemotherapy; age ≤ 20 years, presence of an informal caregiver; and clinic staff and interviewer assessment that the patient had adequate stamina to complete the interview. Patient-caregiver dyads in which either the patient or caregiver refused to participate, met criteria for dementia or delirium (by neuro-behavioral cognitive status exam), or did not speak either English or Spanish were excluded.

Of the 917 eligible patients, 638 patients (69.6%) were consented and enrolled in the larger study. The most common reasons for nonparticipation (N=279) included “not interested” (N=120), “other” (N=69), or “caregiver refuses” (N=37). There were no differences in the socio-demographic characteristics between the participants and non-participants, except that participants were more likely to be Hispanic (11.8% vs. 5.7%, p=0.006). For the present analysis, we restricted our sample to the 332 patients who died to examine the medical care that patients received in the final week of life. The deceased cohort did not differ significantly (p<0.05) by cancer type, psychological distress, or rates of psychiatric disorders. However, as expected, this cohort was more debilitated (e.g., had worse performance status and higher symptom burden) and more likely to have characteristics associated with lower socioeconomic status (e.g., be younger, female, uninsured, and identify as a member of an ethnic minority group).^19,20

Protocol and Measures

Research staff at each site participated in a two-day training session covering the study protocol, medical chart extraction, and techniques for interviewing terminally ill patients and their families. Participants were identified from outpatient clinic. All interviews were conducted in English or Spanish. Patient interviews took an average of 30-45 minutes to complete; caregiver interviews typically lasted 60-75 minutes. Patients and caregivers received $25 as compensation. Upon study enrollment, research staff interviewed the patient and caregiver separately, reviewed the medical record, and confirmed each patient’s diagnosis, treatment, and performance status with the physician.

In the baseline interview patients were asked “Have you and your doctor discussed any particular wishes you have about the care you would want to receive if you were dying?” Responses were coded as “1=yes” and “2=no.” Psychosocial factors identified in the literature as important predictors of EOL care were recorded, including patients’ treatment preferences,¹⁴ advance care planning, acknowledgement that their illness was terminal,^{15, 21} religiousness (patients who reported religion was “very important” were coded as “religious”),²² and the doctor-patient relationship.²³ A close doctor-patient relationship was defined as one where patients trusted and respected their doctor, felt respected and “seen as a whole person,” and were very comfortable asking questions about their care. Socio-demographic characteristics were recorded as reported by the patient. Race/ethnicity was analyzed as a potential determinant of EOL medical care because advanced cancer patients of different racial/ethnic groups have been found to vary in their treatment preferences, advance care planning, and healthcare utilization.^24-27 Race/ethnicity was determined by patient self-report in response to the open-ended question: “What race or ethnicity do you consider yourself to be?” Patients who identified themselves as Hispanic were analyzed as such, regardless of whether they reported White or Black status. Mental health measures included: the Structured Clinical Interview for Diagnostic and Statistical Manual-DSM-IV (SCID),^{28, 29} the Endicott Scale,³⁰ and McGill Quality of Life (MQoL) psychological subscale.³¹ Patients’ functional status and comorbid medical conditions were measured with the Karnofsky score³² and the Charlson Comorbidity Index, respectively.³³ QoL was assessed with the MQoL Index’s physical health, symptom, and social support subscales.³¹ Caregiver interviews included self-reported socio-demographic characteristics and baseline measures of physical and mental health measures, including the SCID and the Medical Outcomes Study 36-item Short Form Survey (MOS SF-36).³⁴

Research staff reviewed the list of study participants weekly with clinic staff at each site. Within two to three weeks of a participants’ death research staff reviewed the medical record to document medical care received. This included specific services that have been defined in the literature as indicators of aggressive medical care, including: admission to an intensive care unit, ventilation, resuscitation, chemotherapy, or use of a feeding tube near death.³⁵ In a post-mortem interview, research staff interviewed the formal/informal caregiver involved in the patients’ care during the final week of life to assess the patient’s QoL near death. Specifically, they asked: “In your opinion, how would you rate the overall quality of the patients’ death/last week of life?” Response items were arranged on a Likert scale from 0 “worst possible” to 10 “best possible.” Patients’ location of death and length of time in hospice was documented.

Bereaved caregivers were interviewed a median of 6.5 months after the patient’s death to assess their bereavement adjustment. We chose this point so that caregivers would be likely to be beyond a state of acute grief,³⁶ but close enough to the death to avoid introduction of recall bias. Interviewers repeated the SCID and MOS SF-36. Caregivers were also asked about their satisfaction and regrets about patients’ EOL care.

Statistical analysis

T-test, Cochran-Mantel-Haenszel, and χ² test statistics were used, as appropriate, to test for significant differences between patients who did or did not report EOL discussions with their physicians at baseline. A Cox proportional hazards model was performed to examine differences in survival between the two groups. A log-rank test was used to determine whether the groups differed significantly in survival. Item non-response rates were less than 10% for all study variables, except for a question that assessed patients’ preferences for life-extending therapy (11.6% responded “don’t know”). To preserve the study sample, missing data were imputed at the median value, except for patients’ preference for life-extending therapy which was imputed by the predicted value from a multivariable logistic regression model.

We used a propensity-score weighting technique to balance characteristics that differed significantly (p<0.10) between the two groups (i.e., recruitment site, Karnofsky score, Charlson Comorbidity Index, MQoL symptom subscale, survival, religiousness). Logistic regression models estimated the odds of reporting an EOL conversation as a function of these characteristics. The predicted probabilities of reporting an EOL conversation (propensity scores) were used to derive individual weights equal to the probability of belonging to the opposite group, making the weighted distribution of characteristics among participants in both groups balanced. ³⁷ Therefore weighted analyses adjusted for potential confounding from measured characteristics associated with EOL discussions. Propensity-score weighted logistic regression was used to estimate the effect of EOL discussions on binary outcomes (e.g., psychiatric disorders), while propensity-score weighted linear regression models estimated the effect of EOL discussions on continuous measures (e.g., QoL). Logistic regression models examining the effect of EOL discussion on medical care received in the last week of life were also adjusted for patients’ treatment preferences, desire for prognostic information, and terminal illness acknowledgement.

Multivariable relationships between patients’ QoL at the EOL and the intensity of medical services received were examined. Adjusted one-way ANOVA models tested the associations between patient QoL near death and the number of aggressive procedures as well as length of hospice stay, controlling for Karnofsky score and survival. The relationship between patients’ QoL near death, the number of aggressive procedures received in the last week of life, and the length of hospice stay was plotted.

Caregiver bereavement outcomes estimated the effects of patient medical care and QoL near death on the surviving caregiver’s likelihood of meeting criteria for psychiatric disorders, health-related QoL, and grief reactions (e.g., regret). In these models, baseline measures of each bereavement outcome (e.g., overall QoL) were included to estimate the effect of the death over and above caregiver baseline levels. Other confounding influences were included in the model if they changed the estimate of the main predictor by >10%, and were retained if they remained significant at a level of p<0.05 controlling for other confounders. Confounders considered included: socio-demographics characteristics (age, sex, race, education, and religion), kinship relationship, length of caregiving, source of report for patient QoL, and religiousness. Statistical analyses were performed with SAS version 9.1 (SAS Institute Inc, Cary, NC). We report two-sided p-values without adjustment for multiple comparisons.

RESULTS

Patient characteristics

The cohort consisted of 332 terminally ill cancer patients who died a median of 4.4 months after enrollment. The socio-demographic, clinical, and psychosocial characteristics of the cohort at baseline are described in Table 1. Overall, 123 of 332 (37.0%) patients reported having EOL discussions with their physicians at baseline.

Table 1.

Participant Characteristics by End Of Life (EOL) Care Discussion with a Physician (N=332)

	Total sample (n=332)	Unadjusted Comparison EOL Discussion			Adjusted Comparisona EOL Discussion
		Yes (n=123)	No (n=209)	P value	Yes (n=123)	No (n=209)	P value
Socio-demographic
Age, mean (SD), y	57.9 (12.2)	57.7 (12.0)	58.0 (12.3)	0.79	57.1 (10.6)	57.1 (13.8)	--- b
Sex (%)				0.68
Male	55.1	36.1	63.9		55.0	55.0	--- b
Female	44.9	38.3	61.7		45.0	45.0
Race/Ethnicityc (%)				0.48			0.91
White, non-Hispanic	63.9	68.3	61.2		61.9	61.9
Black, non-Hispanic	19.9	18.7	20.6		22.0	22.0
Hispanic	14.8	12.2	16.3		15.1	15.1
Asian/Other	1.4	0.8	1.9		1.0	1.0
Married (%)	59.8	64.2	57.2	0.22	62.6	53.4	0.28
Education, mean (SD), y	12.6 (4.0)	12.8 (3.7)	12.5 (4.1)	0.51	12.6 (3.2)	12.6 (4.2)	--- b
Health insurance (%)	57.1	57.0	57.1	0.99	54.2	51.3	0.74
Religion (%)				0.45			0.81
Catholic	36.6	36.6	35.1		35.9	32.8
Protestant	17.2	16.3	17.8		17.3	17.7
Baptist	16.6	15.5	17.3		17.4	19.4
Jewish	3.0	0.8	4.3		0.5	2.5
Muslim	0.9	0.8	1.0		0.9	0.8
Pentecostal	2.7	4.1	1.9		4.4	2.5
Other	18.1	16.3	19.2		16.3	21.1
None	4.8	7.3	3.4		7.4	3.2
Recruitment site (%)				<0.0001			0.86
Yale Cancer Center	22.9	11.0	30.7		13.7	13.7
VACCC	5.4	7.3	4.3		8.0	4.0
Parkland Hospital	44.6	46.3	43.5		50.8	47.4
Simmons Center	8.7	7.6	9.9		7.3	11.5
DFCI/MGH	2.7	2.4	3.0		2.2	4.0
NHOH	15.7	3.5	9.9		16.4	16.4
Clinical characteristics
Cancer type (%)				0.42			0.96
Breast	11.5	13.8	10.1		13.6	12.3
Colorectal	15.1	17.9	13.4		18.8	13.6
Pancreatic	8.7	7.3	9.6		6.9	9.0
Other GI	11.8	10.6	12.4		10.4	13.3
Lung	22.9	17.9	25.8		19.8	19.8
Otherd	30.4	32.5	29.2		30.5	32.1
Performance status mean (SD)
Karnofskye mean (SD)	63.4 (15.4)	59.5 (15.1)	65.7 (15.1)	0.0003	62.1 (13.0)	62.1 (16.9)	--- b
Charlsonf mean (SD)	8.2 (2.6)	8.5 (2.7)	8.0 (2.5)	0.09	8.2 (2.2)	8.2 (2.9)	--- b
McGill physicalg mean (SD)	5.7 (2.6)	5.5 (2.8)	5.8 (2.6)	0.27	5.7 (2.3)	5.6 (3.0)	0.62
McGill symptomg mean (SD)	5.4 (2.1)	5.0 (2.1)	5.6 (2.1)	0.02	5.3 (1.8)	5.3 (2.3)	--- b
Survival, median, d [25%-75% quartiles]	131.5 [59.5-258.5]	95.0 [44.0-211.0]	154.0 [72.0-292.0]	0.002	104.0 [52.0-224.0]	104.0 [50.0-225.0]	0.87
Doctor-patient relationshiph (%)	68.3	71.5	67.0	0.39	73.1	66.5	0.41
Religiousi (%)	69.3	63.4	72.7	0.08	69.3	69.3	--- b
Social supportg	1.7	8.6	8.6	0.86	8.7	8.7	0.92

^aIn adjusted comparisons, responses were weighted according to the estimated propensity score of participants reporting an EOL discussion.

^bBecause there were no differences in the characteristics between groups after propensity score weighting, the p-value is 1.00.

^cRace/ethnicity was determined by patient self-report in response to the open-ended question: “What race or ethnicity do you consider yourself to be?”

^dThe remaining patients had cancer types each representing <5% of the sample. Due to rounding, this totals to more than 100%.

^eKarnofsky score is a measure of functional status that is predictive of survival, where 0 is dead and 100 is perfect health.

^fThe Charlson comorbidity index is an age-adjusted measure of comorbid illness, where higher numbers signify a greater burden.

^gSubscales of the McGill Quality of Life Questionnaire (scale 0-10) where 0 is undesirable and 10 is desirable.

^hPercentage of patients with a “close doctor-patient relationship,” defined as one where patients trusted and respected their doctor, felt respected and “seen as a whole person,” and were very comfortable asking questions about their care.

ⁱPercentage of patients who reported religion was “very important.”

There were no associations between EOL discussions and patients’ self-reported socio-demographic characteristics (age, sex, race, marital status, education, insurance, or religion), cancer type, relationships with physicians, religiousness, or social support. There were, however, significant site differences with 61.5% of patients at New Hampshire Hematology Oncology reporting EOL conversations compared to 16.2% at Yale Cancer Center (p<0.001). Patients reporting discussions had lower performance status (59.5 vs. 65.7; p=0.003), higher symptom burden (5.0 vs. 5.6; p=0.02), and shorter survival times (95 vs. 154 days; p=0.002), compared to patients who did not.

Propensity-score weighting successfully adjusted for significant differences in the baseline characteristics of patients who reported EOL conversations and those who did not. After adjustment participants no longer differed by their Karnofsky score, symptom burden, survival time, recruitment site, or religiousness.

Caregiver characteristics

Caregivers’ socio-demographic characteristics were similar to those of the patients, except that they were significantly (p<0.05) younger (mean age: 51.2 vs. 57.9 years), more likely to be female (77.0% vs. 44.9%), and more highly educated (mean: 13.2 vs. 12.6 years). The caregiver-patient kinship relationships were: spouse (51.4%), adult child (24.0%), other relative (13.9%), friend (6.6%), and parent (4.2%). The median length of time spent caregiving prior to baseline was 11.5 months (interquartile range: 5.0-30.0 months).

Mental health, treatment preferences, and advance care planning

As shown in table 2, EOL discussions were not associated with patients feeling “depressed,” “sad,” “terrified,” “worried,” or meeting DSM-IV-criteria for a mental disorder in propensity-weighted analyses. Patients who reported EOL conversations were significantly (p≥0.001) more likely to accept that their illness was terminal (52.9% vs. 28.7%), prefer medical treatment focused on relieving pain and discomfort over life-extending therapies (85.4% vs. 70.0%), and complete a do-not-resuscitate order (63.0% vs. 28.5%).

Table 2.

Associations Between Advanced Cancer Patients’ End-of-Life Discussions, Mental Health, Terminal Illness Acceptance, Treatment Preferences, and Planning

	Total Sample (n=332)	EOL Discussion		AOR (95% CI)a
		Yes (n=123)	No (n=209)
Mental disorders, N (%)
Major depressive disorderb	22 (6.7)	10 (8.3)	12 (5.8)	1.33 (0.54-3.32)
Major depressive disorder-Endicottc	20 (6.1)	7 (5.8)	13 (6.3)	0.73 (0.26-2.06)
Generalized anxiety disorderb	7 (2.1)	4 (3.3)	3 (1.4)	2.50 (0.51-12.1)
Panic disorderb	10 (3.1)	2 (1.7)	8 (3.9)	0.55 (0.16-1.90)
Posttraumatic stress disorderb	9 (2.7)	4 (3.3)	5 (2.4)	0.95 (0.24-3.75)
Any mental disorderb	33 (10.2)	11 (9.2)	22 (10.7)	0.73 (0.35-1.55)

McGill psychological subscale,d adjusted least square means (SE)	Total Sample	Yes	No	P value
“Depressed”	7.4 (2.9)	7.3 (0.2)	7.4 (0.2)	0.79
“Nervous or worried”	6.9 (3.2)	6.5 (0.3)	7.0 (0.3)	0.19
“Sad”	7.2 (3.0)	7.3 (0.2)	7.2 (0.2)	0.79
“Terrified”	7.2 (3.1)	7.1 (0.3)	7.2 (0.3)	0.68
Any psychological distress	5.4 (0.1)	5.3 (0.2)	5.4 (0.2)	0.55

Acceptance, preferences and planning, N (%)	Total Sample	Yes	No	AOR (95% CI)
Accepts illness is terminal	125 (37.7)	65 (52.9)	60 (28.7)	2.19 (1.40-3.43) ***
Wants to know life expectancy	242 (72.9)	103 (83.7)	139 (66.5)	2.40 (1.43-4.04) **
Values comfort over life-extension	245 (73.8)	105 (85.4)	140 (70.0)	2.63 (1.54-4.49) ***
Against death in ICU	118 (35.5)	60 (48.8)	58 (27.8)	2.13 (1.35-3.37) **
Completed DNR order	134 (41.1)	75 (63.0)	59 (28.5)	3.12 (1.98-4.90) ***
Completed living will, durable power of attorney, or health care proxy	181 (55.2)	86 (71.7)	95 (46.1)	1.96 (1.25-3.07) **

^*P value<0.05

^**P value<0.01

^***P value<0.001

^aAOR=adjusted odds ratio, the propensity-weighted sample was used in these analyses.

^bDiagnosed according to Structured Clinical Interview for DSM-IV for mental disorders.

^cThe Endicott scale has been validated for depression in terminally-ill cancer patients. It measures non-somatic symptoms of depression.

^dSubscales of the McGill Quality of Life Questionnaire (scale 0-10) where 0 is undesirable and 10 is desirable.

Medical care at the end-of-life

In propensity-score weighted analyses, adjusted for patients’ desire for prognostic information, terminal illness acceptance and treatment preferences, patients who reported EOL conversations with their physicians at baseline received significantly fewer aggressive medical interventions near death. As shown in table 3, patients who reported EOL discussions were less likely to receive mechanical ventilation (1.6% vs. 11.0%; p=0.02), undergo resuscitation (0.8% vs. 6.7%; p=0.03), or be admitted to the intensive care unit (4.1% vs. 12.4%; p=0.03). Patients reporting EOL discussions were more likely to be enrolled in outpatient hospice for more than one week (65.6% vs. 44.5%, p=0.03).

Table 3.

Medical Care Received in the Last Week of Life by End of Life (EOL) Discussion (N=332)

	Total	EOL Discussion
		Yes	No	AOR (95%CI)a
Medical care received in the last week, N (%)	332	123 (37.0)	209 (63.0)
ICU admission	31 (9.3)	5 (4.1)	26 (12.4)	0.35 (0.14-0.90)*
Ventilator use	25 (7.5)	2 (1.6)	23 (11.0)	0.26 (0.08-0.83)*
Resuscitation	15 (4.5)	1 (0.8)	14 (6.7)	0.16 (0.03-0.80)*
Chemotherapy	19 (5.7)	5 (4.1)	14 (6.7)	0.36 (0.13-1.03)
Feeding tube	26 (7.9)	11 (8.9)	15 (7.3)	1.30 (0.55-3.10)
Outpatient hospice utilized	213 (64.4)	93 (76.2)	120 (57.4)	1.50 (0.91-2.48)
Outpatient hospice ≤ 1 week	173 (52.3)	80 (65.6)	93 (44.5)	1.65 (1.04-2.63)*

P value<0.05

^**P value<0.01

^***P value<0.001

^aAOR=Adjusted odds ratio. The propensity-score weighted sample was used for these analyses. Logistic regression models were also adjusted for patients’ treatment preferences, desire for prognostic information, and whether they accept that their illness is terminal.

Patients’ Quality of Life at EOL

Patients who received aggressive medical interventions had worse QoL in the final week of life. Patients’ QoL decreased with increasing numbers of aggressive medical therapies, even after adjusting for their severity of illness (see Figure 1). Patients who received no aggressive care had a mean QoL score of 6.4, while patients who received 3 or more therapies had a mean QoL score of 4.6 (F=3.61, df=3, p=0.01). The inverse was true for hospice care, where patients’ QoL improved the longer they were enrolled, except for patients who received less than one week of services. Patients who did not receive hospice had a mean QoL score of 5.6; those enrolled for one week or less had a QoL score of 5.6; and those enrolled for 2 months or more had QoL scores of 6.9 (F=4.04, df=3, p=0.008).

Figure 1.

Results adjusted for illness severity, as measured by Karnofsky score and survival. Hospice length: 0=no hospice, 1=less than one week, 2=one to eight weeks, 3=more than eight weeks (F=4.04, df=3, p=0.008).

Aggressive interventions: 0=none, 1=one intervention (ventilation, resuscitation, chemotherapy, or feeding tube), 2=two interventions, 3=three or more (F=3.61, df=3, p=0.01).

Bereavement Outcomes

In table 4, we display analyses that examined the associations between patients’ experience in the last week of life and caregivers’ bereavement adjustment, controlling for significant confounds (caregivers’ socio-demographics and baseline QoL). Caregivers of patients who received any aggressive care were at higher risk for developing a Major Depressive Disorder (AOR 3.37, 95% CI 1.12-10.13), experiencing regret (β=0.17; p=0.01), and feeling unprepared for the patient’s death (β=−0.30; p<0.001), compared to caregivers of patients who did not receive aggressive care. They also had worse QoL outcomes, including overall QoL (β=−0.15; p=0.004), self-reported health (β=−0.12; p=0.04), and increased role limitations (β=0.17; p=0.008).

Table 4.

Associations Between Caregivers’ Outcomes and Patients’ EOL Care and Quality of Life (QoL) (N=202)

	Aggressive medical care		Hospice care		Patients’ QoL
Caregiver bereavement outcomes	Standardized coefficienta	P value	Standardized coefficienta	P value	Standardized coefficienta	P value
Health-related QoL b
Overall QoL	−0.15c,d	0.004	0.06c,d	0.31	0.20c,e,f	<0.001
Self-reported health	−0.12c	0.04	−0.06c	0.30	0.17c,e	0.004
Physical function	−0.10c,g	0.05	0.010c,g	0.85	0.14c,e,h	0.02
Mental health	−0.11c,i	0.06	0.01c,j	0.81	0.13c,e,j	0.04
Role limitation	0.17c,k	0.008	−0.03c	0.62	−0.10c,e	0.15
Change in health [AOR, (95% CI)]	0.57 (0.29-1.11)c	0.10	1.55 (0.89-2.69)c	0.12	1.17 (1.05-1.29)c,e	0.003
Grief reaction
Felt prepared for death	−0.30c,l	<0.001	0.13c,l	0.05	0.23c,e,l	0.002
Regret	0.17	0.01	−0.08f	0.25	−0.30e	<0.001
Mental disorders m
Any mental disorder [AOR, (95% CI)]	2.25 (0.81-6.23)c,n	0.12	0.87 (0.38-1.97)c,n	0.78	0.87 (0.73-1.03)c,e	0.10
Major depressive disorder [AOR, (95% CI)]	3.37 (1.12-10.13)c,n	0.03	0.46 (0.17-1.21)c,n	0.12	0.86 (0.71-1.05)e	0.09

^aExcept where noted, data are expressed as standardized coefficients which reflect the impact of a change of one standard deviation in the predictor on the outcome variable

^bQuality of life was measured with the 36-item Medical Outcome Study Short-Form Health Survey; Caregiver confounds adjusted for in multivariable models include

^cbaseline measures assessed a median of 10.8 months earlier

^esource of report (formal/informal cargiver) for QoL measure, and the following variables

^dsex and education

^feducation

^gage, education, length of caregiving

^hage and length of caregiving

ⁱsex and relationship with patient (spouse vs. other)

^jsex, education, and relationship with patient (spouse vs. other)

^ksex

^lrace

^mdiagnosed according to Structured Clinical Interview for DSM-IV for mental disorders

ⁿage

As illustrated in Figure 2, there was a direct relationship between patients’ QoL near death and their bereaved caregivers’ QoL at follow-up. High patient QoL was associated with better caregiver outcomes, including overall QoL (β=0.20; p=0.001), self-reported health (β=0.17, p=0.004), physical functioning (β=0.14, p=0.02), mental health (β=0.13, p=0.04), and improvements in self-rated change in health (AOR 1.17, 95% CI 1.05, 1.29). Caregivers of patients with high QoL felt better prepared for the death (β=0.23, p=0.002) and experienced less regret (β=−0.30, p<0.001) at follow-up.

DISCUSSION

Our results suggest that EOL discussions may have cascading benefits for patients and their caregivers. Despite physicians’ concerns that patients may suffer psychological harm due to EOL discussions,^5-7,9 we found no evidence that they were significantly associated with increased emotional distress or psychiatric disorders. Instead, the worst outcomes were seen in patients who did not report EOL conversations. This group received significantly more aggressive medical care in their final week of life, which was associated with worse patient quality of life near death. In addition, their bereaved caregivers experienced worse quality of life, more regret, and were at higher risk for developing a Major Depressive Disorder a median of 6.5 months later.

On the other hand, patients who reported EOL discussions received less aggressive medical care and were more likely to receive hospice services for more than one week. Less aggressive care and earlier hospice referrals were associated with better patient QoL near death. Of note, patients who received less than one week of hospice care had the same QoL scores as patients who did not receive hospice at all, suggesting that patients benefit more from early hospice referrals. Better patient QoL near death, in turn, was associated with better QoL among surviving caregivers who experienced less regret and showed improvements in self-reported health, physical functioning, mental health, and overall QoL during the bereavement period.

To date, most of the communications literature in cancer has focused on doctors’ and patients’ preferences surrounding prognostication with little attention paid to the psychological and medical outcomes of these conversations.³⁸ The bereavement literature has begun to explore the associations between patients’ place of death, the receipt of hospice care, and caregivers’ subsequent risk for psychiatric disorders.^39-40,17 Recent studies have shown that communication interventions in the intensive care unit can reduce psychological distress among bereaved family members.⁴¹ Our results suggest that bereavement-related distress might be offset by interventions aimed at reducing aggressive care.

Our findings must be interpreted within the context of an observational study that could not randomize terminally ill cancer patients to EOL discussions for ethical and logistical reasons. In lieu of a trial, propensity-score weighting enabled us to examine our primary outcome independent of observed differences. Although this technique cannot correct for unmeasured or hidden biases, such as how a patients’ prior experiences with death may influence their attitudes to EOL discussions or decision-making, it is one of the most robust statistical methods available to correct for potential selection bias and confounding. Nevertheless, it is possible that patients who have a preference for less aggressive medical care may be more likely to initiate EOL discussions with their physicians. Alternatively, it is possible that the timing of EOL discussions is important, or that EOL discussions are a proxy for physician factors which may determine medical treatment at the EOL (e.g., physicians who have EOL discussions may limit their patients’ exposure to aggressive medical measures by choosing not to hospitalize patients with terminal illness).

We were further constrained by the limited information available on EOL discussions. For example, we do not know who initiated the conversation, when it happened, or what was said. Our study did not include interviews with doctors or audio-taped conversations. Unfortunately, EOL discussions are often poorly documented in the medical record, and other studies have revealed that patients are caregivers recall of conversations often disagree.⁹ Absent such independent validation, the accuracy of patients’ reported rates of discussions remains unknown, and we suspect our results are a conservative estimate of the true point prevalence of EOL conversations.^9,46 In addition, it worth noting that our sample had disproportionately high rates of ethnic minority patients who were highly symptomatic and had poor performance statuses. Future research is needed to determine whether these findings generalize to patients who are less debilitated and have higher socio-economic status.

One of the strengths of this study is that it includes well validated structured clinical interviews for the diagnosis of mental disorders. Although there was no evidence that EOL discussions were associated with greater psychological distress, it is possible that results may differ in samples with higher rates of psychiatric morbidity. Because our data were cross-sectional, we cannot make causal inferences about patients’ psychological state at the time of the discussion, or their immediate reactions following it. However, if EOL discussions did evoke substantial psychological distress, we would expect to detect a positive association between the two.

The association between EOL discussions and patients’ preference for less aggressive care is noteworthy. EOL discussions may make patients more realistic about the benefits of aggressive therapies, and thereby reduce the likelihood that they receive intensive treatments near death.^15,43 In this study, we cannot confirm the direction of this association because we did not examine changes in preferences resulting from EOL discussions.

In this study, more than 60% of dying patients do not recall having EOL discussions with their physicians. Several possible explanations exist. Cancer patients frequently misunderstand or fail to recall prognostic discussions, even when they occur.^44-47 Physicians also often avoid EOL conversations, communicate euphemistically, are overly optimistic, or delay discussions until patients are close to death, perhaps because their own feelings of failure or loss.^{4,5,10,13,48,49} Consistent with this hypothesis, patients with lower Karnofsky scores, higher symptom burdens, and shorter median survivals were more likely to report having EOL conversations. Interestingly, patients’ recall of EOL discussion varied significantly by site, with fewer patients reporting EOL discussions at a major academic center. Future studies are needed to examine factors influencing patient selection of medical centers for care, and whether centers have distinct cultures that dictate the style and content of EOL conversations.

Given the adverse outcomes associated with not having EOL discussions there appears to be a need to increase the frequency of these conversations. By acknowledging that the EOL is near, patients, caregivers, and physicians can focus on clarifying patients’ priorities and improving pain and symptom management.^1,50